An Efficient Rainfall Prediction Scheme based on Sliding Window Algorithm with Effective Distance Metric Measure

Abstract

Rainfall prediction is essential for the planning and management of water resources and farming. Accurate prediction of rainfall is a difficult task due to the chaotic nature of atmosphere. This paper presents an approach for rainfall prediction using Sliding Window concept with various distance metric measures algorithm.  The Sliding Window Algorithm (SWA) observes the data during the same period in a previous year and predicts rainfall in the following year.  Various distance metrics are used and compared to improve the performance.  Using Sliding Window Algorithm with Effective Distance metric measure (SWA-ED), month wise results are being computed. Based on this algorithm, the rainfall prediction test was conducted for Tirunelveli District, Tamil Nadu, India using the rainfall data for five-year period.